Operating environments: Indoor grasp tests and outdoor okra field trial in Kagoshima, Staged human-robot sparring ring, Staged household-fixture demonstration, Temple of Heaven outdoor choreographed performance…
Control modes: Autonomous, Teleoperation, Shared Control, Scripted, Planner Guided
Auki conference guidance - Demonstrated - 2025-09-09 - Guide visitors to selected conference booths using a pre-mapped Auki domain and onboard sensors. - environment: Pre-mapped WOW Summit conference floor - Software: External Auki network domain with spatial map, points of interest and route planning - Control: autonomous - Conditions: Venue mapped before demonstration - source: auki.com
Dynamic kip-up recovery - Demonstrated - 2025-03-21 - Flip from lying on the back to a standing position in a kip-up demonstration. - environment: Selected indoor dynamic-recovery demonstration - Control: unspecified - Conditions: Live control mode, sensors and compute unspecified - source: youtube.com
Perceptive stepping stones - Research demonstration - 2026-10-13 - Cross four alternating stepping stones using chest-depth perception in a proof-of-concept trial. - environment: Physical sparse-stepping-stone research course - Hardware: Onboard Jetson Orin NX; Chest RealSense D435i - Software: MARCH 50Hz perceptive controller - Control: autonomous - Conditions: Single reported successful proof-of-concept crossing - source: arxiv.org
Multi-terrain recovery - Research demonstration - 2026-07-31 - Recover after a fall and resume commanded locomotion on flat ground, gravel, grass and slopes. - environment: Flat, gravel, grass and inclined physical test surfaces - Hardware: 29-DoF G1 with proprioceptive state input - Software: UniReLo controller, July31 v4 - Control: shared_control - source: arxiv.org
Dynamic locomotion - Research demonstration - 2026-07-30 - Dodged different thrown balls using head-camera depth segmentation, succeeding on 95% of seeded real-world throws. - environment: controlled real-world humanoid dodgeball benchmark - Hardware: fixed head-mounted depth camera; onboard proprioception - Software: PAC-MAN Link-CBF policy; semantic ball segmentation - Control: autonomous - Conditions: seeded single throws; walk-back and recovery between throws - source: arxiv.org
Navigation and obstacle avoidance - Research demonstration - 2026-07-20 - Navigated cluttered indoor scenes without task-specific tuning after simulation-only training. - environment: cluttered real-world indoor scenes - Hardware: Intel RealSense D435i RGB-D camera - Software: perception-only adaptive safety critic and navigation policy - Control: autonomous - Conditions: zero-shot simulation-to-real transfer - source: arxiv.org
Reference-guided beam and box traversal - Research demonstration - 2026-07-15 - Traverse a beam and step onto a box by tracking reference motions from EgoHTR. - environment: Physical beam and box traversal research setup - Hardware: 29-DoF G1; Terrain height-scan input; hardware sensing method unspecified - Software: Per-clip EgoHTR reference-tracking policies - Control: scripted - source: arxiv.org
Range-guided wall bypass - Research demonstration - 2026-06-22 - Bypass wall barriers and recover toward a staged target without continuous joystick steering. - environment: Physical wall-barrier feasibility course - Hardware: Local range-sensing and odometry inputs - Software: LP-NavOA steering policy, locomotion backbone and command-side safety filter - Control: autonomous - source: arxiv.org
Loaded walking with disturbance feedback - Research demonstration - 2026-06-15 - Maintain commanded forward walking while carrying a one-sided hand load in disturbance tests. - environment: Physical loaded-walking and stance research tests - Hardware: Right-hand attached test weights; External RTX5070 Ti Laptop GPU over Ethernet - Software: ADAPT 50Hz policy and dynamics observer; Onboard FAST-LIO 100Hz odometry - Control: shared_control - source: arxiv.org
Depth-guided natural-terrain traversal - Research demonstration - 2026-06-03 - Traverse uneven ground, natural steps and slopes while adapting between walking and running. - environment: Outdoor natural-terrain research tests - Hardware: Onboard RealSense D435i depth camera - Software: CoRe-MoE unified gait and terrain-adaptation policy - Control: shared_control - source: arxiv.org
Multi-terrain locomotion - Research demonstration - 2026-06-02 - Traversed boxes, hurdles, stairs and mixed terrain using onboard perception. - environment: mixed-terrain research course - Hardware: Unitree G1 with onboard perception; back-mounted NVIDIA Jetson Thor; onboard NVIDIA Jetson Orin - Software: terrain-aware diffusion motion generator; reinforcement-learning whole-body motion tracker - Control: planner_guided - source: wholebodylocomotion.github.io
Visual-memory pedestrian and glass avoidance - Research demonstration - 2026-04-01 - Navigate corridors and clutter, wait for doors to open, and avoid pedestrians or glass in supervised research trials. - environment: Indoor clutter, corridor, glass-wall and pedestrian research scenes - Hardware: Chest RealSense T265 and D455; Onboard Jetson Orin NX; Networked Jetson Thor, optionally backpack-mounted - Software: EgoNav diffusion navigation and visual memory; DINOv3 features and segmentation; Receding-horizon trajectory controller - Control: autonomous - Conditions: Supervised trials with recorded human interventions - source: arxiv.org
Commanded running with onboard mini-PC - Research demonstration - 2026-03-26 - Run on a treadmill or follow remotely set outdoor velocity and heading commands. - environment: Indoor treadmill and outdoor sidewalk research trials - Hardware: Robot-powered onboard mini-PC - Software: CLF-RL running controller; FAST-LIO odometry and layered heading/position controller - Control: shared_control - source: arxiv.org
LiDAR-guided running avoidance - Research demonstration - 2026-03-26 - Avoid detected outdoor obstacles while running under a LiDAR-based safety-planning layer. - environment: Outdoor obstacle-avoidance research setup - Hardware: Robot-powered onboard mini-PC; LiDAR local sensing - Software: CLF-RL running controller; Local occupancy map and Poisson safety function; CBF-constrained MPC velocity planner - Control: autonomous - source: arxiv.org
Depth-guided parkour skill selection - Research demonstration - 2026-02-17 - Vault obstacles, climb walls and land drops while a depth policy selects skills along commanded travel directions. - environment: Depth-perceived parkour research obstacle courses - Hardware: 29-DoF G1 with onboard depth camera - Software: PHP depth-conditioned multi-skill visuomotor policy - Control: shared_control - Conditions: High-level navigation uses supplied 2D velocity commands - source: arxiv.org
LiDAR-guided platform traversal - Research demonstration - 2026-02-11 - Climb onto a platform, crawl across it and climb down through geometry-triggered whole-body skills. - environment: Physical high-platform traversal research course - Hardware: 29-DoF G1; Livox MID-360 LiDAR; Intel Core i7 terrain-perception computer - Software: APEX unified traversal policy and elevation-map pipeline - Control: autonomous - Conditions: Traversal direction is commanded; skill transitions use terrain geometry; Motion capture used for trajectory analysis - source: arxiv.org
Skateboard propulsion and steering - Research demonstration - 2026-02-03 - Push a skateboard, steer by leaning and transition between pushing and coasting phases. - environment: Indoor and outdoor skateboard research tests - Hardware: Skateboards in the tested configurations - Software: HUSKY phase-specific learned controller at 50Hz - Control: unspecified - Conditions: Foot-placement guidance and markers constrain contact to the central deck - source: arxiv.org
Dual-depth stair ascent and descent - Research demonstration - 2025-10-16 - Climb and descend a three-step flight using fused chest and hip depth views. - environment: Instrumented stair and compliant-ledge research course - Hardware: Chest and rear-hip Intel RealSense D435 cameras; Onboard Jetson Orin NX 16GB; Framework Laptop 13 / Ryzen7 7840U controller computer - Software: Two-stage depth-conditioned locomotion policy - Control: shared_control - Conditions: Supplied forward velocity command; Controller laptop may be offboard or strapped to robot - source: arxiv.org
Compliant ledge traversal - Research demonstration - 2025-10-16 - Step across a hinged plank and onto a compliant mat at a 36cm ledge. - environment: Instrumented stair and compliant-ledge research course - Hardware: Chest and rear-hip Intel RealSense D435 cameras; Onboard Jetson Orin NX 16GB; Framework Laptop 13 / Ryzen7 7840U controller computer - Software: Two-stage depth-conditioned locomotion policy - Control: shared_control - Conditions: Supplied forward velocity command; Controller laptop may be offboard or strapped to robot - source: arxiv.org
Commanded walk-run transitions - Research demonstration - 2025-09-25 - Transition smoothly between walking and running while following velocity commands up to 2.5 metres per second. - environment: Physical locomotion research trials - Software: RuN conditional motion generator and residual policy - Control: shared_control - source: arxiv.org
Gait-conditioned walking and standing - Research demonstration - 2025-05-27 - Walk, stand and transition from walking to standing under supplied motion commands. - environment: Physical gait-control research tests - Software: Gait-conditioned RL locomotion policy - Control: shared_control - Conditions: Gait IDs derive from commanded velocity and Froude number - source: arxiv.org
Proprioceptive standing-up recovery - Research demonstration - 2025-02-12 - Rise from lying, seated or leaning postures across indoor surfaces and outdoor grass, stone and platforms. - environment: Indoor and outdoor posture-recovery research trials - Hardware: 23-actuated-joint G1; Onboard Jetson Orin NX; IMU and joint encoders - Software: HoST posture-adaptive standing-up policy - Control: autonomous - Conditions: Additional prone/side-lying behaviors appear in later paper revision; Payload disturbance trials do not establish load pickup or transport - source: arxiv.org
Flashsac blind walking - Research demonstration - Walk forward, backward and sideways using a proprioceptive locomotion policy. - environment: FlashSAC physical G1 flat-ground and stair test course - Hardware: 29-DoF G1 research configuration - Software: FlashSAC-trained locomotion policy; CENet proprioceptive history/context estimator - Control: unspecified - Conditions: Velocity-command interface; runtime human command source unspecified; No exteroceptive terrain perception used - source: arxiv.org
Flashsac blind stairs - Research demonstration - Climb stairs with 15-centimetre risers using a blind locomotion policy. - environment: FlashSAC physical G1 flat-ground and stair test course - Hardware: 29-DoF G1 research configuration - Software: FlashSAC-trained locomotion policy; CENet proprioceptive history/context estimator - Control: unspecified - Conditions: Velocity-command interface; runtime human command source unspecified; No exteroceptive terrain perception used - source: arxiv.org
Configured okra grasping - Demonstrated - 2026-09-29 - Detect, reach and grasp okra with a camera-guided right-hand cutter attachment; cutting remained unfinished. - environment: Indoor grasp tests and outdoor okra field trial in Kagoshima - Hardware: Chest ZED-M stereo camera; Right Dex1-1 hand with cutter attachment; left basket; Backpack Jetson Orin 64GB and battery - Software: YOLO okra detection, stereo positioning and inverse kinematics; Diffusion policy for final alignment - Control: unspecified - Conditions: Cutting and force control were unfinished - source: prtimes.jp
Requested package retrieval - Demonstrated - 2026-07-20 - Retrieve a requested coloured medicine package from stacked boxes. - environment: Staged household-fixture demonstration - Software: UnifoLM-OminiA-0.3 - Control: autonomous - Conditions: Manufacturer-described autonomous execution; Hand vendor, sensors and inference location unspecified - source: youtube.com
Two-controller loco-manipulation - Research demonstration - 2026-10-05 - Wipe a sink while crouched, move carts or chairs, and pick up a low box while an operator supplies two hand targets. - environment: Physical whole-body interaction research setup - Hardware: Two hand controllers with per-hand clutch inputs - Software: OCLO onboard analytic posture prior and lower-body policy; Differential IK arms with motor-torque force estimation - Control: shared_control - source: arxiv.org
Reference-tracked suitcase pulling - Research demonstration - 2026-10-05 - Pull a suitcase while walking along the supplied motion reference. - environment: Indoor physical motion-imitation research setup - Hardware: 29-DoF G1, firmware1.5.4; External laptop with RTX5090 Laptop GPU and Ethernet Unitree SDK - Software: InterMimicGen retargeted reference trajectory; SONIC body reference tracking at50Hz; onboard joint PD at500Hz - Control: scripted - Conditions: Object-blind motion tracking; no live object feedback or object replanning; Ankle-pitch stiffness/damping scale1.5; pelvis-IMU pitch correction1degree; Suitcase end-effector model unspecified - source: arxiv.org
Reference-tracked tripod carrying - Research demonstration - 2026-10-05 - Grasp and carry a tripod using Inspire hands synchronized with the body reference. - environment: Indoor physical motion-imitation research setup - Hardware: 29-DoF G1, firmware1.5.4; External laptop with RTX5090 Laptop GPU and Ethernet Unitree SDK; Two Inspire RH56DFTP dexterous hands - Software: InterMimicGen retargeted reference trajectory; SONIC body reference tracking at50Hz; onboard joint PD at500Hz; Recorded finger trajectory replay at about12Hz; body-frame synchronization - Control: scripted - Conditions: Object-blind motion tracking; no live object feedback or object replanning; Ankle-pitch stiffness/damping scale1.5; pelvis-IMU pitch correction1degree - source: arxiv.org
Dual-robot handling with Vega1 - Research demonstration - 2026-09-22 - Clear or nudge objects, tilt a board and receive a doll while coordinating with a Vega1 partner. - environment: Dual-robot collaborative manipulation research setup - Hardware: Dexmate Vega1 partner robot in the studied configuration; G1 RealSense egocentric camera - Software: DUET ACT collaborative policy; SONIC G1 execution controller - Control: autonomous - source: arxiv.org
Chip-bag shelf placement - Research demonstration - 2026-07-22 - Grasp a chip bag from a trolley box, rotate toward a shelf and place it using an offboard learned policy. - environment: Supermarket-replica research setup - Hardware: G1-Edu with Dex-3 hands; Head RealSense D435i and two wrist D405 cameras; Offboard RTX5090 workstation over Ethernet - Software: DEED post-trained GR00T N1.6 policy; Unitree locomotion controller - Control: autonomous - Conditions: Refined policies require manual reset after each placement - source: arxiv.org
Whole-body loco-manipulation - Research demonstration - 2026-07-20 - Completed 71 of 80 trials across eight real-world loco-manipulation task families using persistent 3D object-state verification. - environment: real-world humanoid loco-manipulation evaluations - Hardware: Dex3-1 dexterous hand; head-mounted RGB-D camera; external desktop workstation for perception and task execution; onboard low-level arm and locomotion control - Software: POT-VLA; GR00T-N1.7 whole-body action expert - Control: autonomous - Conditions: object movement; contact; occlusion; recovery - source: arxiv.org
Stationary pear placement - Research demonstration - 2026-06-30 - Pick a pear from the table and place it into a basket. - environment: Stationary tabletop manipulation research setup - Hardware: Brainco Revo2 Tactile hands; Head-mounted RealSense D435i camera; Fixed lower limbs and waist - Software: Task-trained GR00T N1.5 policy with head-image and joint-state observations - Control: autonomous - Conditions: Teleoperation demonstrations train the policy; the retained action is learned-policy evaluation - source: arxiv.org
Learned page turning - Research demonstration - 2026-06-30 - Lift and turn a page until it lies flat on the other side. - environment: Stationary tabletop manipulation research setup - Hardware: Brainco Revo2 Tactile hands; Head-mounted RealSense D435i camera; Fixed lower limbs and waist - Software: Task-trained GR00T N1.5 policy with head-image and joint-state observations - Control: autonomous - Conditions: Teleoperation demonstrations train the policy; the retained action is learned-policy evaluation - source: arxiv.org
Bimanual book insertion - Research demonstration - 2026-06-30 - Hold a document bag with one hand and insert a book with the other. - environment: Stationary tabletop manipulation research setup - Hardware: Brainco Revo2 Tactile hands; Head-mounted RealSense D435i camera; Fixed lower limbs and waist - Software: Task-trained GR00T N1.5 policy with head-image and joint-state observations - Control: autonomous - Conditions: Teleoperation demonstrations train the policy; the retained action is learned-policy evaluation - source: arxiv.org
Learned bottle-cap unscrewing - Research demonstration - 2026-06-30 - Hold a bottle and rotate its cap to unscrew it. - environment: Stationary tabletop manipulation research setup - Hardware: Brainco Revo2 Tactile hands; Head-mounted RealSense D435i camera; Fixed lower limbs and waist - Software: Task-trained GR00T N1.5 policy with head-image and joint-state observations - Control: autonomous - Conditions: Teleoperation demonstrations train the policy; the retained action is learned-policy evaluation - source: arxiv.org
Externally tracked moving-target reaching - Research demonstration - 2026-06-29 - Replan whole-body movement to reach a moving target whose pose is supplied by an external tracker. - environment: Physical moving-target research trials - Hardware: Offboard RTX4090 workstation; HTC Vive Ultimate Tracker on moving target with Wi-Fi pose stream - Software: ReactiveBFM TensorRT planner; 50Hz whole-body tracking controller - Control: autonomous - source: arxiv.org
Vision-guided box handling - Research demonstration - 2026-06-29 - Approach selected objects, turn around them, and pick or place boxes on the floor and raised surfaces. - environment: Lab and apartment research scenes with varied furniture layouts - Hardware: Head-mounted ZED2i on custom 3D-printed bracket; Tethered RTX5000 Ada laptop; External RTX5090 inference server - Software: VLK policy trained from synthetic reconstructed-scene interactions; Websocket inference client and 50Hz whole-body tracker - Control: autonomous - source: arxiv.org
Trajectory tracking - Research demonstration - 2026-06-24 - Tracked asynchronous upper-body task-space trajectories on the physical humanoid. - environment: humanoid-control research laboratory - Hardware: 29-DoF Unitree G1 research configuration - Software: asynchronous sparse upper-body task-space tracking policy; high-level sparse trajectory planner - Control: planner_guided - source: arxiv.org
Motion-capture-guided ball contact - Research demonstration - 2026-06-24 - Hit forehand and backhand tennis shots and contact a soccer ball using externally tracked targets. - environment: Motion-capture-equipped physical research setup - Hardware: 27-DoF G1; Eight-camera 120Hz OptiTrack system over Ethernet; Tennis racket for tennis trials - Software: TaskNPoint onboard policy with 50Hz target updates - Control: shared_control - source: arxiv.org
Motion-capture-guided box handling - Research demonstration - 2026-06-24 - Pick up and place a box using TaskNPoint targets and external motion capture. - environment: Motion-capture-equipped physical research setup - Hardware: 27-DoF G1; Eight-camera 120Hz OptiTrack system over Ethernet - Software: TaskNPoint onboard policy with 50Hz target updates - Control: shared_control - source: arxiv.org
Dex3 recorded-trajectory replay - Research demonstration - 2026-06-22 - Replay walking-and-grasping motions and pull a mock refrigerator door while stepping backward. - environment: Qualitative physical trajectory-replay research setup - Hardware: G1 with Dex3-1 hands; Simplified refrigerator-door mock-up for OpenFridge - Software: CoorDex recorded joint-position trajectories and low-level replay interface - Control: scripted - source: arxiv.org
Language-conditioned bin placement - Research demonstration - 2026-06-19 - Select a red ball, cup, bottle or medicine bottle from language input and place it in a shared bin. - environment: Shared-bin tabletop research setup - Hardware: Two Inspire RH56DFTP dexterous hands; Head RealSense D435i - Software: VQActFlow learned policy and whole-body control; TWIST2 for training-demonstration collection - Control: autonomous - source: arxiv.org
Active-neck grasp and transfer - Research demonstration - 2026-06-17 - Pick and transfer bread, bags or towels, including grasping while crouched, using an active camera neck. - environment: Physical active-perception manipulation research setup - Hardware: Robot-mounted Pika grippers with gripper cameras; Custom three-DoF active camera neck - Software: HALOMI pi0.5 high-level policy; Manifold-constrained 50Hz head-hand whole-body controller - Control: autonomous - source: arxiv.org
Restricted-envelope arm reaching - Research demonstration - 2026-06-14 - Move the left arm to sampled reachable Cartesian targets using an energy-aware policy. - environment: Physical restricted-envelope arm-reaching setup - Software: Frozen energy-aware arm policy; Python ROS2 control node at 100Hz; Reachability-screened Cartesian targets and torque safety clamp - Control: autonomous - source: arxiv.org
Camera-guided whole-body handling - Research demonstration - 2026-06-08 - Push and load a cart, raise a basket, retrieve a bag and stock objects on shelves through whole-body motion. - environment: Physical whole-body manipulation research setup - Hardware: Dual ALOHA2 grippers; Head-mounted RealSense D435i RGB camera; External RTX4090 policy workstation - Software: MotionWAM policy server over WebSocket; SONIC whole-body execution controller - Control: autonomous - source: arxiv.org
VLM-directed whole-body task sequences - Research demonstration - 2026-06-04 - Pick, carry and place objects, including crouched and two-arm sequences, using the tested perception-and-planning stack. - environment: Physical language-directed manipulation research setup - Hardware: ZED-M RGB-D camera; Jetson Thor - Software: HANDOFF 50Hz controller; VLM/depth perception with deterministic waypoint, skill and gripper tools; Tested local VLM and ChatGPT API configurations - Control: autonomous - source: arxiv.org
Dex1 world-model manipulation - Research demonstration - 2026-06-03 - Open a pot lid and place a potato inside, select a red bottle, or place an object on a marked target. - environment: Physical tabletop model-evaluation setup - Hardware: Dex1-1 grippers - Software: Flash-WAM distilled policy configurations - Control: autonomous - source: arxiv.org
Object manipulation - Research demonstration - 2026-05-29 - Grasp gaze-selected objects and hand them to a person using voice-triggered human target selection. - environment: interactive manipulation research laboratory - Hardware: Meta Aria eye-tracking glasses worn by the operator; robot-mounted RealSense D455 RGB camera - Software: Gaze2Act gaze-conditioned policy; GR00T N1.5 - Control: shared_control - source: arxiv.org
Pose-tracked bottle grasping - Research demonstration - 2026-04-19 - Track and grasp a drink bottle at five desk positions, then lift it with a Dex3 hand. - environment: Restricted tabletop bottle-grasping research workspace - Hardware: Dex3 dexterous hands; RealSense D435i RGB-D camera; External RTX4060 Laptop GPU platform - Software: YOLOv8 and FoundationPose object tracking; Unity inverse-kinematics trajectory planner; C++ UDP bridge and Unitree SDK - Control: autonomous - source: arxiv.org
XL-VLA bimanual tasks - Research demonstration - 2026-04-15 - Prepare fruit, pass a bottle between hands, and pour sauce or sugar using the XL-VLA policy with Inspire hands. - environment: Tabletop bimanual research setup - Hardware: G1 with Inspire dexterous hands - Software: XL-VLA latent-action policy; HOMIE and MANUS gloves for demonstration collection - Control: autonomous - source: manus-meta.com
Active-camera cube placement - Research demonstration - 2026-03-30 - Find small cubes placed across the workspace and grasp and place them into a box. - environment: Stationary tabletop sensory-ablation research setup - Hardware: G1 Edu-U4 with Dex3-1 three-finger hands; Active stereo camera; additional sensor combinations vary by evaluated policy - Software: Augmented LeRobot ACT policies - Control: autonomous - source: arxiv.org
Image-conditioned static-scene actions - Research demonstration - 2026-03-10 - Move a box, sit on a sofa or chair, or kick a ball after generating a motion plan from the initial view. - environment: Mostly static scene-aware whole-body research trials - Hardware: Chest-mounted GoPro camera; External RTX4090 server; G1 onboard tracking computer - Software: ZeroWBC fine-tuned Qwen-VL motion generation; Motion decoder, retargeting and onboard interactive tracker - Control: autonomous - Conditions: Open-loop generated plan; no replanning after initial observation; Target object or structure must be visible initially - source: arxiv.org
MoCap-guided object-motion tracking - Research demonstration - 2026-03-03 - Lift or carry boxes and transport a suitcase while tracking dense motion references with external object poses. - environment: OptiTrack-equipped object-transport research setup - Hardware: External OptiTrack object-pose system - Software: ULTRA onboard student policy; Dense object/body motion references - Control: scripted - Conditions: Two physical trials per reported configuration - source: arxiv.org
Depth-guided operator goal following - Research demonstration - 2026-03-03 - Move boxes toward operator-specified future poses using onboard depth observations and the learned controller. - environment: Physical sparse-goal object-manipulation research setup - Hardware: Onboard depth sensing for egocentric point clouds; Keyboard goal-command interface - Software: ULTRA onboard student policy; Depth point-cloud extraction and sparse object-transform goals - Control: shared_control - Conditions: External MoCap used for evaluation even when policy uses onboard depth; Two physical trials per reported configuration - source: arxiv.org
Compliant surface contact - Research demonstration - 2026-01-21 - Follow a curved surface and press a held massage ball against a tabletop object. - environment: Tabletop contact-control research trials - Hardware: Two 1-DoF grippers without wrist force-torque sensing; Head-mounted Intel RealSense D435i; External RTX4090 / Intel i9-13900K workstation; Onboard PC impedance controller at 50Hz - Software: HumanoidVLM Molmo-7B-O visual task inference; FAISS retrieval of impedance and gripper parameters; Task-space virtual impedance control and inverse kinematics - Control: unspecified - Conditions: Visual retrieval selects control gains; reference task trajectories are supplied; No wrist force-torque feedback; reported forces are virtual proxies - source: arxiv.org
Asymmetric egg and bottle placement - Research demonstration - 2026-01-21 - Place an egg and a sauce bottle with separate compliance settings for each arm. - environment: Tabletop contact-control research trials - Hardware: Two 1-DoF grippers without wrist force-torque sensing; Head-mounted Intel RealSense D435i; External RTX4090 / Intel i9-13900K workstation; Onboard PC impedance controller at 50Hz - Software: HumanoidVLM Molmo-7B-O visual task inference; FAISS retrieval of impedance and gripper parameters; Task-space virtual impedance control and inverse kinematics - Control: unspecified - Conditions: Visual retrieval selects control gains; reference task trajectories are supplied; No wrist force-torque feedback; reported forces are virtual proxies - source: arxiv.org
Fork contact and object lifting - Research demonstration - 2026-01-21 - Poke produce with a held fork and grasp and lift tabletop objects. - environment: Tabletop contact-control research trials - Hardware: Two 1-DoF grippers without wrist force-torque sensing; Head-mounted Intel RealSense D435i; External RTX4090 / Intel i9-13900K workstation; Onboard PC impedance controller at 50Hz - Software: HumanoidVLM Molmo-7B-O visual task inference; FAISS retrieval of impedance and gripper parameters; Task-space virtual impedance control and inverse kinematics - Control: unspecified - Conditions: Visual retrieval selects control gains; reference task trajectories are supplied; No wrist force-torque feedback; reported forces are virtual proxies - source: arxiv.org
GR00T household object retrieval - Research demonstration - 2025-12-15 - Approach household fixtures to rack a mug, retrieve a snack or eraser, box a shoe and place an apple in a basket. - environment: Staged indoor household-fixture research setup - Hardware: Parallel grippers shown in the native clips; Trailing cable visible in the physical test setup - Software: Task-post-trained GR00T N1.6; Real-time chunking during asynchronous policy rollout - Control: autonomous - Conditions: Hand vendor, sensor model and inference-compute location unspecified - source: research.nvidia.com
Rubber-hand cube contact and transfer - Research demonstration - 2025-10-13 - Approach a cube to press or push it, and transfer a cube between hands. - environment: Object-pose-tracked indoor loco-manipulation research setup - Hardware: Custom 2-DoF Dynamixel neck with RealSense D435; Rubber hands for contact interaction - Software: DemoHLM learned manipulation policy; FoundationPose++ object-pose tracking; AMO whole-body controller - Control: autonomous - Conditions: Policies trained using simulation-generated demonstrations; Known-object pose tracking; unmodeled-object manipulation not established - source: arxiv.org
Parallel-gripper cube pickup - Research demonstration - 2025-10-13 - Approach a cube, grasp it with the custom parallel grippers and lift it. - environment: Object-pose-tracked indoor loco-manipulation research setup - Hardware: Custom 2-DoF Dynamixel neck with RealSense D435; Custom 3D-printed parallel grippers with Dynamixel actuation - Software: DemoHLM learned manipulation policy; FoundationPose++ object-pose tracking; AMO whole-body controller - Control: autonomous - Conditions: Policies trained using simulation-generated demonstrations; Known-object pose tracking; unmodeled-object manipulation not established - source: arxiv.org
Parallel-gripper cabinet opening - Research demonstration - 2025-10-13 - Reach for a cabinet handle and pull it open with a custom gripper. - environment: Object-pose-tracked indoor loco-manipulation research setup - Hardware: Custom 2-DoF Dynamixel neck with RealSense D435; Custom 3D-printed parallel grippers with Dynamixel actuation - Software: DemoHLM learned manipulation policy; FoundationPose++ object-pose tracking; AMO whole-body controller - Control: autonomous - Conditions: Policies trained using simulation-generated demonstrations; Known-object pose tracking; unmodeled-object manipulation not established - source: arxiv.org
Learned chair rotation - Research demonstration - 2025-10-09 - Rotate a chair with the configured Dex3-1 hands. - environment: Dex3-1 tabletop manipulation research setup - Hardware: G1 with Dex3-1 hands; Intel RealSense RGB-D sensing - Software: Task-trained imitation-learning policy through network policy server - Control: autonomous - Conditions: Human resets between evaluations; autonomous scene reset not established; Inference tasks and teleoperation dataset are separate - source: arxiv.org
Learned tabletop towel folding - Research demonstration - 2025-10-09 - Fold a towel laid on a desk. - environment: Dex3-1 tabletop manipulation research setup - Hardware: G1 with Dex3-1 hands; Intel RealSense RGB-D sensing - Software: Task-trained imitation-learning policy through network policy server - Control: autonomous - Conditions: Human resets between evaluations; autonomous scene reset not established; Inference tasks and teleoperation dataset are separate - source: arxiv.org
Learned toy placement and handover - Research demonstration - 2025-10-09 - Place a dumpling toy into a plate or hand it to a person. - environment: Dex3-1 tabletop manipulation research setup - Hardware: G1 with Dex3-1 hands; Intel RealSense RGB-D sensing - Software: Task-trained imitation-learning policy through network policy server - Control: autonomous - Conditions: Human resets between evaluations; autonomous scene reset not established; Inference tasks and teleoperation dataset are separate - source: arxiv.org
Language-commanded tabletop arrangement - Research demonstration - 2025-10-06 - Pick up a bottle or toy and arrange fruit and toys relative to other tabletop objects. - environment: Apartment and tabletop language-guided research setup - Hardware: Adaptive grippers; Head-mounted downward RGB-D camera; Chest-mounted forward RGB-D camera - Software: Humanoid-COA; GPT-4V scene descriptions and GPT-4 action planning; IK manipulation and learned locomotion primitives - Control: autonomous - Conditions: October6 revised source and inspected native project video; Compute location and camera models unspecified - source: arxiv.org
Language-commanded bottle relocation - Research demonstration - 2025-10-06 - Carry a Coke bottle through the apartment and place it on the requested counter. - environment: Apartment and tabletop language-guided research setup - Hardware: Adaptive grippers; Head-mounted downward RGB-D camera; Chest-mounted forward RGB-D camera - Software: Humanoid-COA; GPT-4V scene descriptions and GPT-4 action planning; IK manipulation and learned locomotion primitives - Control: autonomous - Conditions: October6 revised source and inspected native project video; Compute location and camera models unspecified - source: arxiv.org
Motion-capture-guided door traversal - Research demonstration - 2025-09-20 - Push a door open by hand, walk through and return by kicking it open. - environment: Motion-capture-instrumented whole-body interaction research setup - Hardware: External motion-capture system; Markers on robot pelvis and manipulated objects - Software: HDMI task-specific learned whole-body controller - Control: autonomous - Conditions: Policy input includes tracked pelvis and door poses - source: arxiv.org
Language-selected object placement - Research demonstration - 2025-06-11 - Place the requested fruit, soap or snack packet onto a plate among tabletop distractors. - environment: Language-conditioned tabletop robot-policy research setup - Software: Post-trained GR00T N1.5 policy - Control: autonomous - Conditions: Task-specific post-training on 1,000 teleoperation episodes - source: research.nvidia.com
Teleoperated fruit placement - Research demonstration - 2025-06-01 - Pick apples, pears, starfruit or grapes and place them on a plate under upper-body teleoperation. - environment: Tabletop fruit-placement data collection - Hardware: Default RealSense camera; G1 tri-finger hands - Software: GR00T teleoperation data-collection setup - Control: teleoperation - source: huggingface.co
Wired learned tomato handling - Research demonstration - 2025-02-18 - Squat to pick a tomato from a low sofa and pick and place a tomato using a learned policy. - environment: Wired multi-camera learned-manipulation research setup - Hardware: Dex3 hands; Head RealSense D455 and two arm D435 cameras; External RTX4080 host over Ethernet - Software: Task-trained Seer visuomotor policy; HOMIE whole-body controller - Control: autonomous - Conditions: Fifty teleoperation demonstrations per task train the autonomous policy - source: arxiv.org
Model-controlled towel folding - Research demonstration - Fold a towel on a tabletop using the evaluation’s dexterous hand configuration. - environment: Tabletop towel-folding research setup - Hardware: Task-specific dexterous hands shown in the associated Unitree G1 evaluation clips; Additional camera fixtures shown in the evaluation setup - Software: UnifoLM-WLA-1.0 - Control: autonomous - Conditions: Published clips labelled 2x playback - source: unigen-x.github.io
Model-controlled plug insertion - Research demonstration - Insert a plug into a tabletop power strip in the configured evaluation. - environment: Tabletop plug-insertion research setup - Hardware: Task-specific dexterous hands shown in the associated Unitree G1 evaluation clips; Additional camera fixtures shown in the evaluation setup - Software: UnifoLM-WLA-1.0 - Control: autonomous - Conditions: Published clips labelled 2x playback - source: unigen-x.github.io
Inspire teleoperated washer loading - Research demonstration - Lift clothing from a pile and load it into a front-loading washing machine under human control. - environment: Household-fixture teleoperation data-collection setup - Hardware: Inspire dexterous hands; Head stereo and wrist cameras - Control: teleoperation - Conditions: Human-controlled collection; no learned autonomous rollout established by this dataset - source: huggingface.co
Dex3-1 intended object handling - Claimed - 2024-05-16 - Handle eggs, open soda bottles and manipulate hard objects with the optional Dex3-1 hands. - environment: Manufacturer-described configured manipulation applications - Hardware: G1 EDU with optional Dex3-1 hands - Control: unspecified - Conditions: Specific task tool and execution control unspecified - source: prnewswire.com
Fifty-robot martial-arts formation - Demonstrated - 2026-02-23 - Perform synchronized leaps, shifts, flips and a fist-clasping salute in a 50-robot formation. - environment: Temple of Heaven outdoor choreographed performance - Hardware: Fifty G1 robots in the demonstrated group - Control: unspecified - Conditions: Choreographed group performance; control architecture unspecified - source: news.cgtn.com
Dual-G1 contact coordination - Research demonstration - 2026-03-03 - Synchronize hugs, greetings and contact routines with a second G1 using shared pose and timing information. - environment: Pre-mapped dual-humanoid interaction research setup - Hardware: Two G1 robots with onboard internal CPU control; LiDAR and IMU sensing; Wireless inter-robot communication - Software: Rhythm ONNX policy at 50Hz; POINT-LIO, GICP and pre-built point-cloud localization map; LCM peer-pose and phase synchronization - Control: autonomous - source: arxiv.org
Language-commanded apartment navigation - Research demonstration - 2025-10-06 - Approach a named sofa, fridge or television and stop near the target. - environment: Apartment and tabletop language-guided research setup - Hardware: Adaptive grippers; Head-mounted downward RGB-D camera; Chest-mounted forward RGB-D camera - Software: Humanoid-COA; GPT-4V scene descriptions and GPT-4 action planning; IK manipulation and learned locomotion primitives - Control: autonomous - Conditions: October6 revised source and inspected native project video; Compute location and camera models unspecified - source: arxiv.org
World-model-controlled sparring - Demonstrated - 2026-09-07 - Step, punch and kick while sparring with a human in a staged ring demonstration. - environment: Staged human-robot sparring ring - Hardware: Boxing gloves on both hands - Software: UnifoLM-X2-1.0 world-action model - Control: autonomous - Conditions: Manufacturer-described autonomy in a selected demonstration; detailed sensing and compute unspecified - source: youtube.com
Motion-capture avatar teleoperation - Demonstrated - 2025-11-05 - Mirror a motion-capture-suit operator to kick a ball, box and perform household tidying motions. - environment: Full-body avatar demonstration and data-collection setup - Hardware: Operator motion-capture suit and straps - Software: Unitree Embodied Avatar full-body teleoperation platform - Control: teleoperation - Conditions: Human operator supplies the task motions - source: youtube.com
Staged athlete fighting interaction - Demonstrated - 2025-08-28 - Perform fighting motions while interacting with athletes at the UFC Performance Institute. - environment: Staged demonstration at UFC Performance Institute - Software: Motion training informed by CHINGMU optical capture of human athletes - Control: unspecified - Conditions: July2025 occurrence known to month; evidence date is dated issuer recap; No live robot motion-capture, onboard CHINGMU hardware or autonomy claim established - source: en.chingmu.com
Robot-versus-robot kickboxing - Demonstrated - 2025-05-31 - Punch and kick another G1 in a staged boxing ring. - environment: Televised staged robot kickboxing ring - Hardware: Boxing gloves - Control: unspecified - Conditions: Control mode unspecified in the retained report - source: forbes.com
Reference motion replay - Research demonstration - 2026-07-22 - Replay locomotion, crouching and dancing references while maintaining balance during manual pushes. - environment: Physical motion-tracking research setup - Hardware: 29-DoF G1; Cabled Intel Core Ultra 7 165H laptop - Software: YAHMP policy with ROS2 at 50Hz - Control: scripted - source: arxiv.org
Optical motion-capture teleoperation - Research demonstration - 2026-06-03 - Track an operator’s dense body references or sparse hand targets derived from optical motion capture. - environment: Optical motion-capture-equipped research setup - Hardware: Optical human motion-capture system - Software: M3imic dense-reference or sparse-end-effector policy - Control: teleoperation - source: arxiv.org
IMU whole-body teleoperation - Research demonstration - 2026-05-12 - Follow an operator’s walking, sit-stand, turning, bowing and arm gestures through IMU retargeting. - environment: Physical whole-body teleoperation tests - Hardware: Virdyn full-body IMU suit - Software: Direct kinematic retargeting controller - Control: teleoperation - source: arxiv.org
Dex3 whole-body teleoperation - Research demonstration - 2026-03-31 - Retarget an operator’s finger, wrist and body inputs to dexterous manipulation and locomotion commands. - environment: Humanoid demonstration-data collection setup - Hardware: Dex3-1 hands; PICO VR headset, wrist, waist and foot trackers; MANUS gloves - Software: PSI teleoperation framework and RL lower-body controller - Control: teleoperation - source: manus-meta.com
Externally generated language motions - Research demonstration - 2026-03-17 - Execute walking, squatting, kicking or hand-gesture references generated from language on an external server. - environment: Physical language-motion tracking research setup - Hardware: 29-DoF G1 EDU; Networked external motion-generation server - Software: ECHO text-to-motion server over WebSocket; Onboard distilled motion tracker - Control: scripted - source: arxiv.org
Teleoperated endoscope positioning - Research demonstration - 2026-02-27 - Position a rigid endoscope under continuous VR control while a surgeon performs a cadaver procedure. - environment: Cadaveric endoscopic-surgery research operating room - Hardware: Karl Storz Image1 HD endoscope and camera system; Custom rigid right-arm 3D-printed endoscope adapter; Meta Quest3 headset and handheld controllers - Software: Mixed-reality teleoperation and inverse-kinematics control - Control: teleoperation - Conditions: Cadaveric proof of concept with a human surgeon and trained teleoperator; No force feedback; continuous verbal coordination - source: arxiv.org
Language-generated whole-body pantomime - Research demonstration - 2026-01-19 - Mime an elephant or violin playing, and take steps before miming a phone call. - environment: Physical text-to-motion research demonstrations - Hardware: G1 EDU+ with Dex3-1 hands; Waist pitch and roll locked; Onboard Jetson Orin NX 16GB; External motion-generation workstation - Software: FRoM-W1 H-GPT motion generation; H-ACT controller through RoboJudo - Control: autonomous - Conditions: Mimed actions from supplied language instructions; Global localization uses ZED Mini or Unitree SDK odometry - source: arxiv.org
Assisted staircase gesture sequence - Research demonstration - 2025-09-20 - Climb stairs, bow, sit, wave and jump down in a sequence with occasional human support. - environment: Motion-capture-instrumented whole-body interaction research setup - Hardware: External motion-capture system; Markers on robot pelvis and manipulated objects - Software: HDMI task-specific learned whole-body controller - Control: shared_control - Conditions: Human operators occasionally support the robot at the back during stair climbing - source: arxiv.org
Locked-wrist whole-body motion tracking - Research demonstration - 2025-07-10 - Track supplied whole-body gestures, steps and dance-like motion sequences with the wrist joints locked. - environment: Whole-body motion-tracking research tests - Hardware: 29-DoF G1 with all six wrist joints locked - Software: UniTracker CVAE motion-tracking policy; Retargeted offline motion references - Control: autonomous - Conditions: Reference-motion tracking; external task planning not established - source: arxiv.org
Exoskeleton-controlled handling - Research demonstration - 2025-02-18 - Move objects between low and high shelves, open an oven and operate a handheld barcode scanner under cockpit control. - environment: Full-body manipulation teleoperation research setup - Hardware: Dex3 hands; Isomorphic exoskeleton arms, motion-sensing gloves and foot pedal; Head-mounted D455 camera; External CPU host and onboard locomotion computer - Software: HOMIE cockpit and locomotion controller - Control: teleoperation - source: arxiv.org
ASAP dynamic motion tracking - Research demonstration - 2025-02-03 - Track kicking, forward-jump, single-foot balance and celebration-motion references. - environment: Instrumented dynamic whole-body motion research trials - Software: ASAP-fine-tuned motion-tracking policies; Locomotion controller between motion sequences - Control: autonomous - Conditions: Motion capture records calibration/evaluation data; Task-specific reference motions; no visual scene planning - source: arxiv.org
Standing side flip - Demonstrated - 2025-04-02 - Launch from standing, rotate sideways in the air and land back on the feet. - environment: Selected indoor acrobatics demonstration - Software: Reinforcement-learning-trained side-flip controller - Control: unspecified - Conditions: Live control trigger and compute configuration unspecified - source: youtube.com
Westlake-model ballet sequence - Demonstrated - 2025-03-21 - Turn on one leg, perform two airborne kicks and finish with a stable bow in a ballet demonstration. - environment: Selected physical dance demonstration - Software: Westlake Robotics embodied motion model - Control: unspecified - Conditions: No remote control and no speedup are attributed company claims; Runtime sensors, inference hardware and reference/trigger interface unspecified - source: hznews.hangzhou.com.cn
Perception-guided soccer shooting - Research demonstration - 2026-06-09 - Aim and kick a soccer ball using onboard perception and a marked target. - environment: Indoor and outdoor soccer research tests - Hardware: 29-DoF G1; Head Livox MID-360 and chest RealSense D435; AprilTag target - Software: RoboNaldo ONNX/DDS controller at 50Hz - Control: autonomous - source: arxiv.org
Camera-guided soccer kick - Research demonstration - 2026-06-08 - Kick a soccer ball into a goal net using the MotionWAM whole-body policy. - environment: Physical whole-body manipulation research setup - Hardware: Dual ALOHA2 grippers; Head-mounted RealSense D435i RGB camera; External RTX4090 policy workstation - Software: MotionWAM policy server over WebSocket; SONIC whole-body execution controller - Control: autonomous - source: arxiv.org
Motion-capture-guided rope skipping - Research demonstration - 2026-06-06 - Coordinate rope turning with another humanoid or human, including single human and humanoid jumpers. - environment: Motion-capture-equipped long-rope research setup - Hardware: 29-DoF G1 robots; Optical MoCap with eight rope markers plus robot and player markers - Software: Marope hierarchical multi-agent controller; External control terminal and DDS rhythm/state signals - Control: autonomous - Conditions: External motion capture is required - source: arxiv.org
Omni-modal pillow and clothing tidying - Demonstrated - 2026-07-20 - Pick up a pillow for the sofa and sort clothing into a laundry basket. - environment: Staged household-fixture demonstration - Software: UnifoLM-OminiA-0.3 - Control: autonomous - Conditions: Manufacturer-described autonomous execution; Hand vendor, sensors and inference location unspecified - source: youtube.com
Omni-modal dishwasher loading - Demonstrated - 2026-07-20 - Load a plate into a dishwasher in a staged household demonstration. - environment: Staged household-fixture demonstration - Software: UnifoLM-OminiA-0.3 - Control: autonomous - Conditions: Manufacturer-described autonomous execution; Hand vendor, sensors and inference location unspecified - source: youtube.com
Material handling - Research demonstration - 2026-10-06 - Place boxes on shelves by hand and with a kick, recovering after pushes or box removal in Workhorse evaluations. - environment: Indoor box-sorting research setup - Hardware: Unitree G1 with onboard camera and proprioception - Software: Workhorse - Control: autonomous - Conditions: Human pushes and removal of the manipulated box - source: hybridrobotics.github.io
Reference-controlled box lift - Research demonstration - 2026-07-22 - Stand from a crouch while lifting a pre-held 1.5-kilogram box through a reference motion. - environment: Physical motion-tracking research setup - Hardware: 29-DoF G1; Cabled Intel Core Ultra 7 165H laptop - Software: YAHMP policy with ROS2 at 50Hz - Control: scripted - Conditions: Box already held before replay; no object-state feedback or replanning - source: arxiv.org
Camera-guided whiteboard wiping - Research demonstration - 2026-06-08 - Wipe a whiteboard while coordinating leg, torso and arm motion. - environment: Physical whole-body manipulation research setup - Hardware: Dual ALOHA2 grippers; Head-mounted RealSense D435i RGB camera; External RTX4090 policy workstation - Software: MotionWAM policy server over WebSocket; SONIC whole-body execution controller - Control: autonomous - source: arxiv.org
Camera-guided laundry and waste placement - Research demonstration - 2026-06-08 - Place clothes into a washing machine and throw waste into a bin in configured research tasks. - environment: Physical whole-body manipulation research setup - Hardware: Dual ALOHA2 grippers; Head-mounted RealSense D435i RGB camera; External RTX4090 policy workstation - Software: MotionWAM policy server over WebSocket; SONIC whole-body execution controller - Control: autonomous - source: arxiv.org
Programmed window-glue application - Research demonstration - 2026-04-19 - Move a glue stick along a car quarter-window edge using a configured trajectory template. - environment: Car-window glue-application research setup - Hardware: Dex3 dexterous hands; RealSense D435i RGB-D camera; External RTX4060 Laptop GPU platform - Software: YOLOv8 and FoundationPose object tracking; Unity inverse-kinematics trajectory planner; C++ UDP bridge and Unitree SDK - Control: scripted - source: arxiv.org
Active-camera can sorting - Research demonstration - 2026-03-30 - Grasp beverage cans at known locations and sort them by colour using a configured ACT policy. - environment: Stationary tabletop sensory-ablation research setup - Hardware: G1 Edu-U4 with Dex3-1 three-finger hands; Active stereo camera; additional sensor combinations vary by evaluated policy - Software: Augmented LeRobot ACT policies - Control: autonomous - source: arxiv.org
Rubber-hand box lifting - Research demonstration - 2025-10-13 - Walk to a box and lift it from the floor using the configured rubber hands. - environment: Object-pose-tracked indoor loco-manipulation research setup - Hardware: Custom 2-DoF Dynamixel neck with RealSense D435; Rubber hands for contact interaction - Software: DemoHLM learned manipulation policy; FoundationPose++ object-pose tracking; AMO whole-body controller - Control: autonomous - Conditions: Policies trained using simulation-generated demonstrations; Known-object pose tracking; unmodeled-object manipulation not established - source: arxiv.org
Parallel-gripper board erasing - Research demonstration - 2025-10-13 - Hold an eraser in a custom gripper and wipe a board. - environment: Object-pose-tracked indoor loco-manipulation research setup - Hardware: Custom 2-DoF Dynamixel neck with RealSense D435; Custom 3D-printed parallel grippers with Dynamixel actuation - Software: DemoHLM learned manipulation policy; FoundationPose++ object-pose tracking; AMO whole-body controller - Control: autonomous - Conditions: Policies trained using simulation-generated demonstrations; Known-object pose tracking; unmodeled-object manipulation not established - source: arxiv.org
Learned desk erasing - Research demonstration - 2025-10-09 - Move a held eraser across the desk surface. - environment: Dex3-1 tabletop manipulation research setup - Hardware: G1 with Dex3-1 hands; Intel RealSense RGB-D sensing - Software: Task-trained imitation-learning policy through network policy server - Control: autonomous - Conditions: Human resets between evaluations; autonomous scene reset not established; Inference tasks and teleoperation dataset are separate - source: arxiv.org
Motion-capture-guided load relocation - Research demonstration - 2025-09-20 - Kneel, lift and relocate a suitcase, box or stack of foam mats. - environment: Motion-capture-instrumented whole-body interaction research setup - Hardware: External motion-capture system; Markers on robot pelvis and manipulated objects - Software: HDMI task-specific learned whole-body controller - Control: autonomous - Conditions: Manipulated objects carry motion-capture markers - source: arxiv.org
Model-controlled home tidying - Research demonstration - Move clothes toward a washing machine, bag household waste and tidy kitchen items during model evaluations. - environment: Staged home-task evaluation interiors - Hardware: Task-specific dexterous hands shown in the associated Unitree G1 evaluation clips; Additional camera fixtures shown in the evaluation setup - Software: UnifoLM-WLA-1.0 - Control: autonomous - Conditions: Published clips labelled 1x playback - source: unigen-x.github.io
Model-controlled shelf placement - Research demonstration - Carry objects and place them on shelves in the manufacturer’s whole-body evaluation. - environment: Shelving research setup - Hardware: Task-specific dexterous hands shown in the associated Unitree G1 evaluation clips; Additional camera fixtures shown in the evaluation setup - Software: UnifoLM-WLA-1.0 - Control: autonomous - Conditions: Published clips labelled 1x playback - source: unigen-x.github.io
Model-controlled phone packaging - Research demonstration - Arrange a phone and its packaging in a tabletop model evaluation. - environment: Tabletop phone-packaging research setup - Hardware: Task-specific dexterous hands shown in the associated Unitree G1 evaluation clips; Additional camera fixtures shown in the evaluation setup - Software: UnifoLM-WLA-1.0 - Control: autonomous - Conditions: Published clips labelled 2x playback - source: unigen-x.github.io
Model-controlled conveyor sorting - Research demonstration - Pick building blocks from a moving conveyor and place them in tabletop bins. - environment: Tabletop conveyor-sorting research setup - Hardware: Task-specific dexterous hands shown in the associated Unitree G1 evaluation clips; Additional camera fixtures shown in the evaluation setup - Software: UnifoLM-WLA-1.0 - Control: autonomous - Conditions: Published clips labelled 2x playback - source: unigen-x.github.io
Dex1 tabletop wiping - Research demonstration - Wipe material across a tabletop with a cloth held in a Dex1 gripper. - environment: Tabletop manipulation data-collection setup - Hardware: Dex1 grippers; Head and wrist cameras - Control: unspecified - Conditions: Data-collection recording; autonomous policy execution not established - source: huggingface.co
Dex1 motor-part assembly recording - Research demonstration - Grasp and align motor parts at a tabletop assembly station. - environment: Tabletop manipulation data-collection setup - Hardware: Dex1 grippers; Head and wrist cameras - Control: unspecified - Conditions: Data-collection recording; autonomous policy execution not established - source: huggingface.co
Brainco teleoperated dish loading - Research demonstration - Pick a plate beside the sink and place it into an open dishwasher under human control. - environment: Household-fixture teleoperation data-collection setup - Hardware: Brainco dexterous hands; Head stereo and wrist cameras - Control: teleoperation - Conditions: Human-controlled collection; no learned autonomous rollout established by this dataset - source: huggingface.co
Intended configured welding - Claimed - 2024-05-16 - Perform welding operations with a task-equipped G1 EDU configuration. - environment: Manufacturer-described configured manipulation applications - Hardware: G1 EDU with optional Dex3-1 hands - Control: unspecified - Conditions: Specific task tool and execution control unspecified - source: prnewswire.com